most comfortable with data on fertility rates, mortality rates, and migration rates, and does not customarily work with attitudinal data. Unfortunately, this has not prevented many demographers from making inferences about the attitudes of individuals and groups. Clearly, of high priority in the study of the political, as well as the social, effects of population change is the rigorous collection of data on the ways in which different ages, classes, and ethnic groups perceive population change.
Fourth, we must be more precise concerning the kinds of political effects which might result from population changes. Within a political system, there are at least two types of political effects which seem to be most apparent. One type concerns the political and administrative strains produced by changes in the size, composition, and distribution of the population. For example, marked increases in the number of young people or the number of older people can have great effects on the kinds of services which governments may be asked to provide. An increase in rural density resulting in a fragmentation of landholdings may have an effect on agricultural policy. Or an increase in family size among poor people supported by welfare programs may have effects upon government expenditures. In short, new demands resulting from population changes may affect the size and character of the bureaucracy and the kinds of resources needed by government to meet those demands. A second major type of political effect is on the internal distribution of power. Since in every society-especially,but not exclusively, in democratic societies—the distribution of political power is affected by the proportions of age groups, ethnic groups, or social classes, any changes affecting those numbers, whether from differential fertility rates or migration, can affect the balance in a federal system; for example, the changes in ratio between Hindus and Muslims in northeastern India, Catholics and Protestants in Northern Ireland, and blacks and whites in American cities have had dramatic political consequences.
Fifth and finally, not only must we be more precise about the political effects of demographic change; it is also necessary that we be more precise about the demographic variables as well. There are at least five specific types of population changes which appear to have political effects. Four of these changes are largely the consequence of increases in population, and the fifth is primarily a consequence of economic and political factors, though it, too, may be affected by population growth. These five are: (a) changes in the age structure of a population, (b) changes in family size, (c) changes in the size and density of a population, (d) differential population growth rates among different social classes and ethnic groups, and (e) migration.
EMPIRICAL APPROACHES TO THE STUDY OF POLITICAL DEMOGRAPHY
The rest of this essay will deal with each of these five aspects of population chanee and consider how each mav be related to different tvoes of